Global retention models in reversed-phase liquid chromatography. A tutorial

Abstract

Developing methods in liquid chromatography for complex samples with hundreds of constituents presents significant challenges, particularly when standards are unavailable or unknown, as is the case of natural products such as medicinal plants. Even when all standards are accessible, optimising experimental conditions for effective separation would require extensive, time-consuming experimentation, often impractical in real-world scenarios. To overcome these challenges, we recently introduced and validated a novel approach based on global retention modelling. This approach has been successfully applied not only to complex samples, but also to simpler ones where standards are available. Global retention models differentiate between solute-specific retention parameters and those characterising the column and solvent, which are shared across the entire set of analytes. These common parameters are derived from chromatographic data for a subset of compounds tracked across experiments in the training design. Once the initial model is built, it can be extended to include additional analytes outside the training set, significantly reducing the need for further extensive experiments. This tutorial provides a comprehensive background on global models, along with a step-by-step explanation of the novel approach. To illustrate its practical application, an example is presented involving a large set of diverse compounds, using a MATLAB set of functions created for this tutorial to showcase the implementation of global retention modelling.